Consequences of flexible employment at labour market entry for early career development in the Netherlands

Abstract

In this article it is investigated to what extent flexible employment at labour market entry negatively affects the early career of school-leavers in the Netherlands in the period 1986–2008. Additionally, the question is raised to what extent educational differences exist in this relationship. To answer these questions, Dutch panel data from the OSA Labour Supply Panel are analysed. The results correspond with previous findings from studies on other European countries, supporting the stepping-stone rather than the entrapment hypothesis. A flexible start in the Dutch labour market increases the likelihood of repeated flexible employment and unemployment in the early career, and coincides with less occupational status development and income growth. However, the detrimental effects of flexible employment at labour market entry are only temporary and diminish after some years. Furthermore, no evidence is found for the existence of educational differences in the negative effects of flexible employment at labour market entry.

Keywords

Early career flexible employment the Netherlands temporary employment youth

Introduction

Consequences of flexible employment at labour market entry

Educational qualifications are crucial for a successful entry in the labour market, which in turn determines to a great extent early career development and labour market chances later in life (Bratberg and Nilsen, 2000; Kogan and Müller, 2003). In many European societies, however, the transition from education to work has been far from smooth over the past three decades, mainly due to the occurrence of persistently high (youth) unemployment (Müller and Gangl, 2003). High unemployment rates paved the way for employment flexibility in these countries and, as a consequence, standard employment relationships began to unravel and various types of flexible work arrangements (such as temporary jobs and on-call employment) emerged (Kalleberg et al., 2000). Such flexible work arrangements are usually considered as marginal by workers, as they offer little labour market perspective and security in terms of a stable employment contract with a fixed income, especially compared to permanent jobs. As young workers represent a very vulnerable labour market group, flexible and insecure employment tends to be pronounced especially at labour market entry (Blossfeld et al., 2008).

In the literature, the consequences of flexible employment at labour market entry are regarded from opposing perspectives. On the one hand, quite a large body of literature concerns the negative impact of flexible employment at the beginning of one’s career, that is that people might get ‘trapped’ in such unstable jobs if they once accept one (Scherer, 2005; Steijn et al., 2006). According to this entrapment scenario, the negative consequences of a flexible start go beyond the duration of that specific job and are long lasting. On the other hand, it is argued that a flexible labour market entry, unlike unemployment, offers individuals the opportunity to actively participate in the labour market and to gain work experience, keeping one’s knowledge and skills up to date, at least to some extent. This perspective considers the negative consequences of a flexible job to be temporary, meaning that the later career is not negatively affected. Elaborating on this entry port or integration scenario, it is argued that a flexible entry position may even bring relative advantages for the subsequent career, which is also known as the stepping-stone perspective (Scherer, 2004). The argumentation is that initial disadvantages, due to a flexible labour market entry, ask for greater mobility steps later on. As opposed to the entrapment scenario, flexible employment at labour market entry may hence function as a step towards more permanent employment in the further career (de Graaf-Zijl et al., 2011).

As being outlined above, the two scenarios both assume that the consequences of flexible employment at labour market entry are equal among different groups of young workers. Taking account of the decisive role of education in the allocation of young people to jobs, however, gives rise to the assumption that the consequences of a flexible entry may vary between individuals with different credentials (see Gebel, 2010). As Breen (1997) and Kalleberg (2009) argue, employment flexibility disproportionately affects labour market entrants with low (vocational) qualifications. Previous research on the Dutch case showed that the trend towards labour market flexibilization indeed is different for lower and higher educated entrants (de Lange et al., 2012). It was found that the increasing likelihood to start in a flexible job instead of a permanent one particularly applied to lower educated people, while the growing likelihood to enter flexible employment instead of being unemployed particularly pertained to higher educated people. Whereas a flexible entry for the lower educated hence signals an unstable start in the labour market, for higher educated people it may be more an indication of having had the opportunity to enter the labour market, instead of remaining unemployed.

With regard to the career consequences of flexible employment at labour market entry, a similar differential effect of education may be expected. It has already been mentioned that the perspectives on the early career consequences of a flexible entry diverge in the literature, from being negative to even positive. Following the different implications of a flexible labour market entry for lower and higher educated individuals, it seems legitimate to expect that the negative scenario may be more applicable to the lower educated, while a less negative or even positive scenario may particularly apply to the higher educated.

Research questions

This article investigates the early career effects of flexible employment at labour market entry of young workers in the Netherlands in the period 1986–2008. The first research question answers whether a flexible entry in the labour market negatively affects the early career of Dutch school-leavers and whether this predicted negative effect is lasting or not. Flexible labour market entry is defined here as being in temporary employment in the first job. So, the focus is restricted to numerical flexibility. The early career is studied from a comprehensive perspective by concentrating on changes in the employment situation, occupational status development and income growth. The latter two outcomes are indicators of employment quality (in terms of social standing and remuneration) and thus important to include, in addition to a measure of the actual employment situation of young workers. Previous research (Gash, 2008; McGinnity et al., 2005; Scherer, 2004; Wolbers, 2008) has studied these various outcomes, but not at the same time. The second research question relates to the issue whether educational differences exist in the expected negative effects of a flexible entry in the labour market on the early career of Dutch school-leavers. These educational differences form a rather unexplored area within the literature on the consequences of flexible employment at labour market entry, although they are often suggested (see for an exception Gebel, 2010).

To answer the two research questions, the OSA Labour Supply Panel is used. This panel study, which offers detailed information on the employment history of the (potential) Dutch labour force, started in 1985, with subsequent surveys that have taken place every two years since 1986. On the basis of this panel study, school-leavers were selected who left daytime education in the period 1986–2008. The use of such a prospective, longitudinal research design is quite common in the international literature on the early career effects of a flexible labour market entry among young workers (see for instance McGinnity et al., 2005), but not yet applied in studies on the Netherlands.

The Dutch context

The Netherlands provides an interesting context for this empirical analysis. As a result of the high (youth) unemployment rates of the 1980s, a number of active policy measures have been adopted since then to make the Dutch labour market more flexible. The Wassenaar Agreement of 1982 is considered as the basis for these initiatives and regarded as one of the pillars of the ‘Dutch Miracle’ (Visser and Hemerijck, 1997). As a consequence, flexible employment has increased remarkably in the Netherlands. This has manifested itself in the considerable number of jobs with temporary contracts and the amount of employment through temporary work agencies or on an on-call basis, especially among youth. Between 1992 and 2005, the share of Dutch labour market entrants in flexible employment increased from 22% to 34% (de Lange et al., 2012). Part-time employment is, as a matter of fact, not regarded as flexible employment in the Netherlands, unlike many other countries. In the Dutch case, many part-time jobs are permanent positions, are voluntarily chosen, and protect against unfair dismissal in the same way as full-time jobs (Remery et al., 2002). In fact, the huge increase of part-time work is closely related to the late and rapid arrival of married women into the Dutch labour force and the lack of sufficient childcare provisions (Visser, 2002), rather than being a means of reducing (youth) unemployment.

In addition, the Dutch labour market is characterized by relatively strong employment protection legislation (EPL), resulting in closed employment relationships and relatively powerful interference of labour unions in the negotiation on collective labour agreements, which typically classifies the Netherlands as a coordinated economy (OECD, 2004). Moreover, temporary workers are mostly not unionized, due to the insecure and short-term character of temporary jobs (Goslinga and Sverke, 2003). This implies that the dichotomy between insiders and outsiders in the Dutch labour market is rather distinct. School-leavers, as outsiders in the labour market, may hence struggle to become an insider, and it may take a while before a permanent employment position is obtained after a flexible labour market entry.

Theory and hypotheses

Negative consequences of flexible employment at labour market entry

The entrapment scenario, predicting a negative, long-lasting effect of flexible employment at labour market entry on further career development, is based on ideas from different theories. According to labour market segmentation theory, the labour market is divided into a primary and secondary segment, mainly differing in terms of stability characteristics (Doeringer and Piore, 1971). The primary segment is characterized by jobs with relatively high status and wages, good working conditions, job security and opportunities to move up. Jobs in the secondary segment, conversely, have low status, poor wages and working conditions, and often lack job security and career opportunities (see also Kalleberg et al., 2000). Temporary jobs particularly are examples of jobs in the secondary segment, whereas standard, permanent jobs belong to the primary segment. It is argued that mobility between the primary and secondary segment is hardly possible. People starting in the less favourable segment hence barely have a chance to move to the primary segment. From this point of view, entering the labour market in flexible employment implies being in an unstable, insecure labour market situation for a longer time, possibly even the whole career, and having to change jobs often. As wages are low and chances for upward mobility scarce in the secondary segment, entering in a flexible job also entails (enduring) low wages and low job status.

According to signalling theory (Spence, 1973), employers use certain signals of future employees in their recruiting decisions, in case relevant information (such as work experience) is absent. Flexible employment at labour market entry, in this respect, can be a signal to employers that the person involved lacks some relevant work capacities. The flexible entry job hence works as a stigma from which a future employee might suffer the rest of his or her career, for instance by repeatedly getting fixed-term contracts or jobs with lower occupational status and wages. Being persistently in such a marginal labour market situation may also make an employee less confident in finding a stable job in the future. This could complicate an ‘escape’ from the secondary segment even more.

Lastly, human capital theory (Becker, 1964) assumes that flexible employment at labour market entry negatively affects the opportunities to develop new working skills and knowledge, and may even lead to a loss of productive skills and a lack of work experience. Flexible entrants are thus less attractive for employers, because they need more training (and hence require more investments) to catch up with standard entrants. Employers will ‘compensate’ this lack of human capital by offering lower skilled jobs and paying lower wages.

To summarize, the theories outlined above all lead to the entrapment hypothesis that flexible employment at labour market entry negatively affects the early career of school-leavers in the Netherlands (H1).

Why the negative consequences of flexible employment at labour market entry diminish

Flexible employment at labour market entry may be unfavourable for the subsequent career, as mentioned above, but is this effect equally strong during the early career? Temporary contracts can be considered as a kind of extended probationary period, which can serve as a screening device for employers (Wang and Weiss, 1998). As labour market entrants lack relevant work experience, it is hard for employers to value their skills and productivity. Before being offered a permanent contract, a young worker gets the opportunity to prove his or her work capacities in the temporary job. If the performances meet the expectations of the employer, the temporary contract may be converted into a permanent one. In addition, it is likely that initial lower occupational status and wages, associated with a temporary entry job, converge to the wage and status of people starting in permanent employment, after successful screening. The ‘penalty’ of flexible employment at labour market entry may hence be strongest in the very beginning of the early career and diminish afterwards, or might even disappear. The stepping-stone hypothesis reads, accordingly, that the negative effects of flexible employment at labour market entry on the early career of school-leavers in the Netherlands diminish after a few years in the labour market (H2).

Educational heterogeneity in the negative consequences of flexible employment at labour market entry

There are arguments for expecting a heterogeneous effect of flexible employment at labour market on the early career along the line of educational credentials. It is argued, for instance, that particularly socially deprived groups of young workers are ‘victims’ of employment flexibility (Breen, 1997; Kalleberg, 2009). Due to a process of skill-biased technological change, the current labour market is characterized by more highly skilled jobs (Katz and Autor, 1999). Such jobs are hard to submit to direct supervision, because of the specialized knowledge of the employee holding the job (Breen, 1997). This explains why employers are more willing and forced to offer higher educated long-term employment contracts, and transfer risks and uncertainties in the labour market to the more easily replaceable lower educated. So, employment flexibility disproportionately affects those who already have a weaker position in the labour market. Accordingly, it is expected that lower educated flexible entrants run higher risks of repeated flexible employment in the early career.

In addition, it is very plausible that the nature of flexible employment among lower and higher educated entrants differs. Labour market segmentation theory is based on the assumption that there is a link between the allocation mechanisms in the various labour market segments and the required skills. Since a minimum qualification level is required to gain access to the primary segment of the labour market, it can be expected that lower educated labour market entrants are mainly found in the secondary segment. As mobility between both segments is hardly possible, it is plausible that lower educated young workers who entered the secondary labour market segment are likely to be entrapped in that segment, implying that the adverse effects of flexible employment at labour market entry on the subsequent career are stronger for lower educated than for higher educated labour market entrants (Gebel, 2010). Flexible jobs among the higher educated, conversely, might be more a conscious choice, and are not necessarily of low quality (de Jong et al., 2009). Therefore, it is expected that the negative effects of flexible employment at labour market entry on the early career are smaller for higher educated school-leavers than for lower educated school-leavers in the Netherlands (H3).

Previous evidence

Previous evidence on the impact of flexible employment at labour market entry on the early career of young workers has been mixed. For (West) Germany, various authors (Gash, 2008; Gebel, 2010; McGinnity et al., 2005; Scherer, 2004) found that starting in flexible employment does not seem to clearly signal a ‘bad start’, as it does not significantly reduce one’s career chances. For instance, McGinnity et al. (2005) showed that, after five years in the labour market, the unemployment rates of those starting in temporary employment do not significantly differ from those starting in permanent jobs. Gebel (2010) obtained similar results: temporary entrants experience higher wage penalties and risks of repeated temporary employment, but these effects diminish after five years. In addition, he concluded that, in Germany, the effects of temporary employment at labour market entry appear to be strongest for tertiary educated, suggesting the integration perspective among this group in particular.

The results on other European countries seem to provide evidence for the stepping-stone hypothesis as well. Data on France, Italy, Denmark and Great Britain revealed that a flexible labour market entry has no enduring negative effects on the subsequent career (Gash, 2008; Scherer, 2004). Only in Great Britain do very short job spells appear to lead to a lower status position later on (Scherer, 2004).

A study on Poland showed that both the entrapment and stepping-stone perspective cannot be convincingly supported (Baranowska et al., 2011). In Poland, fixed-term contracts at labour market entry seem to function as a screening device, helping employers to identify the best workers, providing at least some evidence for the stepping-stone hypothesis.

For Spain, it was observed that flexible employment at labour market entry coincides with the experience of subsequent job instability (Iannelli and Soro-Bonmati, 2003), in particular among lower skilled individuals. Furthermore, low transition rates into permanent contracts were found for Spain (Golsch, 2003). These findings lend support for the entrapment scenario.

Also for the Netherlands, the consequences of starting in temporary employment for the subsequent career seem to be negative, and – although decreasing over time – relatively long lasting, which corroborates the entrapment hypothesis (Steijn et al., 2006; Wolbers, 2010). These conclusions, however, are based on studies using either a cross-sectional research design (Wolbers, 2010) or a retrospective one (Steijn et al., 2006), which could involve problems of causality and recall bias, respectively. In this article, it is aimed to get insight into a more comprehensive picture of the early career effects associated with a flexible entry in the Dutch labour market, by using a prospective research design. More specifically, the purpose is to find out to what extent empirical evidence for the entrapment hypothesis is still found, as earlier findings for the Netherlands suggest. Additionally, the possible heterogeneous effect of education is taken into account.

Data and measurements

Data and selection of sample population

For the empirical analysis, the OSA Labour Supply Panel was used. This panel study, providing detailed information on (changes in) the labour market position of the Dutch (potential) workforce, started in 1985 and has been repeated every two years since 1986. In each survey, about 4500 people aged 16 through 64 and not being a full-time student are interviewed, together representing over 2000 households. In case a panel member does not wish to or is not able to participate in later surveys, this person is replaced by newly selected panel members and/or households, who resemble the original participants as much as possible, with regard to characteristics as age, sex, family size or region of residence. In this article, data are used from 1988 until 2008.

School-leavers are defined as individuals who indicated not being in full-time education in one interview (t), whereas they were two years earlier (t-2). Although the exact moment of leaving daytime education is unknown, this is the most accurate estimate that can be made. All school-leavers left school between 1986 and 2008. In order to focus on young workers, individuals older than 34 at the time of leaving education are excluded from the analysis,¹ as well as people in military service after leaving education. The early career is defined as the first eight years in the labour market after leaving daytime education. In addition, only people being employed at the first moment of interview have been investigated.²

The early career is studied by changes in the employment situation, occupational status development and income growth. The employment situation refers to the situation two, four, six and eight years after labour market entry, resulting in an analytical sample of 473 respondents. Occupational status and income are studied from the first observation point, that is the situation at labour market entry and the situation two, four, six and eight years after entry. This inclusion of the first labour market situation results in a larger analytical sample, that is of 973 and 899 respondents, respectively.

Table 1 presents a statistical description of all variables used in the empirical analysis.

Table 1.

Descriptive statistics of the independent and dependent variables.

	N	Distribution	Range	Mean	SD
Dependent variables (level 1)
Labour market situation
Standard employment	901	87.35%
Flexible employment	901	9.32%
Unemployment	901	3.30%
Occupational status	1831		16–88	44.55	14.17
Standardized (log) hourly wage	1621		−3.76–8.04	0.01	0.97
Independent variables (level 1)
Duration	1831		0–8	1.72	2.24
Unemployment rate	1831		1.86–7.72	5.27	1.75
Independent variables (level 2)
Flexible employment (first job)
Standard job	973		0–1	0.71
Flexible job	973		0–1	0.29
Labour market entry cohort
1986–1989	973		0–1	0.17
1990–1993	973		0–1	0.25
1994–1997	973		0–1	0.24
1998–2001	973		0–1	0.13
2002–2008	973		0–1	0.21
Sex
Male	973		0–1	0.47
Female	973		0–1	0.53
Ethnicity
Native	973		0–1	0.97
Non-native	973		0–1	0.03
Level of education
Low	973		0–1	0.31
Intermediate	973		0–1	0.44
High	973		0–1	0.25
Occupational status (first job)	473		16–88	43.82	13.17
Occupational status unknown (first job)	473		0–1	0.05
Type of industry (first job)
Traditional primary/classical capitalist industry	973		0–1	0.05
Competitive industry	973		0–1	0.12
Large-scale engineering based industry	973		0–1	0.05
Small competitive industry	973		0–1	0.29
Professional service industry	973		0–1	0.21
Bureaucratic service industry	973		0–1	0.07
Other industry or unknown	973		0–1	0.21
Firm size (first job)	973		1–10,000	214.63	723.40
Firm size unknown (first job)	973		0–1	0.08

Source: Labour Supply Panel (1988–2008).

Dependent variables

Employment situation is divided into standard employment (a job with a permanent contract, or a temporary job with prospect [provided by the employer] of a permanent contract³), flexible employment (a job with a temporary contract without prospect of a permanent contract, temporary agency work and on-call employment) and unemployment. As indicated above, this variable is measured two years (t+2), four years (t+4), six years (t+6) and eight years (t+8) after labour market entry, and both flexible employment and unemployment are contrasted with permanent employment. Occupational status is measured according to the International Socio-Economic Index (ISEI) of Ganzeboom et al. (1992), ranging from 10 to maximally 90. Income is measured as the log net hourly wage. As wages usually grow over the years, due to currency inflation, the log net hourly wages are standardized within years. Both occupational status and income are measured at labour market entry (t) and every two years after (until eight years after labour market entry).

Independent variables

Several independent variables are included in the analysis. First of all, this regards flexible employment at labour market entry, which distinguishes flexible first jobs from standard ones.⁴ Duration refers to the time passed (in years) since labour market entry. It is a linear variable⁵ and varies between two and eight years after labour market entry with regard to the analysis of changes in the employment situation. Regarding the analysis of occupational status development and income growth, it ranges from the time at labour market entry through eight years after. Level of education is classified in three distinct categories: lower educated (those with primary [LO], lower secondary general [MAVO] or lower secondary vocational [LBO] education), intermediate educated (those with intermediate secondary general [HAVO], higher secondary general [VWO] or intermediate secondary vocational [MBO] education) and higher educated people (those with higher vocational [HBO] or university [WO] education). Both duration and level of education are included as time-varying covariates.

Control variables

As control variables, sex (indicating females vs males) and ethnicity (indicating non-natives vs natives) are included. Labour market entry cohort refers to the year of leaving daytime education, grouped into the following categories: 1986–1989 (reference category), 1990–1993, 1994–1997, 1998–2001 and 2002–2008.⁶ In addition, the current aggregate unemployment rate is included, based on figures from Statistics Netherlands (CBS, 2011).

To assess the effects of flexible employment at labour market entry on the early career of young workers, several job and company characteristics of the first job are statistically controlled for. This way, selection into less or more prosperous labour market sectors may be filtered out, which could subsequently affect early career development. Occupational status of the first job, which is also based on the ISEI, is only included in the analysis on changes in the employment situation. Type of industry of the first job is measured according to the classification of Stinchcombe (1979), distinguishing between traditional primary or classical capitalist industries, competitive industries (reference category), large-scale engineering based industries, small competitive industries, professional service industries and bureaucratic service industries. A category of other industry or unknown type of industry has been added. Finally, firm size related to the first job is included as a linear variable, to control for the fact that temporary entrants in external labour markets (prevailing in small companies) possibly have worse career opportunities (Baron and Bielby, 1984).

Results

Flexible labour market entry and changes in the employment situation

Changes in the employment situation two, four, six and eight years after labour market entry are considered in Tables 2a and 2b. As all the observations (level-1 units) are nested within respondents (level-2 units), two-level multinomial logit models have been estimated, where both flexible employment and unemployment are contrasted with standard employment (the reference category). In Model 1, flexible employment at labour market entry, duration since labour market entry and level of education are included, statistically controlled for sex, ethnicity, labour market entry cohort and current unemployment rate. In Model 2, three first job characteristics are added: occupational status, type of industry and firm size. In Model 3, the statistical interaction term between flexible employment at labour market entry and duration is also included.⁷ Lastly, the interactions between flexible employment at labour market entry and level of education are added in Model 4.

Table 2a.

Repeated measures of labour market situation (as from two years after labour market entry); flexible (1) vs standard employment (0) (N_observations = 901; N_respondents = 473).

	Model 1		Model 2		Model 3		Model 4
	B	SE	B	SE	B	SE	B	SE
Intercept	−4.11**	0.95	−4.11**	1.14	−4.35**	1.16	−4.27**	1.19
Flexible employment (first job)
Standard job	ref.		ref.		ref.		ref.
Flexible job	1.61**	0.28	1.58**	0.30	2.41**	0.59	2.25**	0.75
Duration	−0.05	0.08	−0.04	0.08	0.05	0.09	0.05	0.09
Labour market entry cohort
1986–1989	ref.		ref.		ref.		ref.
1990–1993	0.71	0.37	0.76	0.39	0.76*	0.39	0.76	0.40
1994–1997	−0.20	0.57	−0.07	0.59	−0.11	0.59	−0.12	0.60
1998–2001	0.75	0.69	0.52	0.73	0.53	0.73	0.53	0.74
2002–2006	1.26*	0.59	0.99	0.65	0.99	0.65	0.97	0.66
Sex
Male	ref.		ref.		ref.		ref.
Female	−0.04	0.28	0.02	0.31	0.02	0.31	0.02	0.31
Ethnicity
Native	ref.		ref.		ref.		ref.
Non-native	0.66	0.66	0.56	0.72	0.54	0.73	0.52	0.73
Level of education
Low	ref.		ref.		ref.		ref.
Intermediate	0.00	0.33	0.01	0.36	−0.02	0.36	−0.07	0.44
High	0.23	0.37	0.38	0.46	0.33	0.46	0.23	0.55
Unemployment rate	0.17	0.12	0.16	0.12	0.16	0.12	0.16	0.12
Occupational status (first job)			−0.01	0.01	−0.01	0.01	−0.01	0.01
Type of industry (first job)
Traditional primary/classical capitalist industry			−0.65	0.70	−0.05	0.70	−0.04	0.70
Competitive industry			ref.		ref.		ref.
Large-scale engineering based industry			0.34	0.73	0.29	0.73	0.31	0.73
Small competitive industry			0.27	0.53	0.21	0.53	0.21	0.54
Professional service industry			−0.03	0.61	−0.06	0.61	−0.05	0.61
Bureaucratic service industry			−0.08	0.80	−0.06	0.80	−0.05	0.80
Other industry or unknown			0.60	0.60	0.58	0.60	0.58	0.60
Firm size (first job)			−0.00	0.00	−0.00	0.00	−0.00	0.00
Flexible job*Duration					−0.26	0.16	−0.26*	0.16
Flexible job*Intermediate							0.19	0.71
Flexible job*High							0.27	0.76
Variance	1.09**	0.39	1.30**	0.41	1.28**	0.41	1.32**	0.42

p < .01; *p < .05 (two-sided). Coefficients of ‘occupational status (first job) missing’ and ‘firm size (first job) missing’ not shown.

Source: Labour Supply Panel (1988–2008).

Table 2b.

Repeated measures of labour market situation (as from two years after labour market entry); unemployment (1) versus standard employment (0) (N_observations = 901; N_respondents = 473).

	Model 1		Model 2		Model 3		Model 4
	B	SE	B	SE	B	SE	B	SE
Intercept	−6.11**	1.39	−5.28**	1.65	−5.98**	1.75	−5.49**	1.79
Flexible employment (first job)
Standard job	ref.		ref.		ref.		ref.
Flexible job	1.62**	0.42	1.59**	0.44	3.61**	0.96	3.15**	1.11
Duration	0.06	0.11	0.06	0.11	0.26	0.14	0.24	0.14
Labour market entry cohort
1986–1989	ref.		ref.		ref.		ref.
1990–1993	−0.03	0.52	0.04	0.55	0.08	0.55	−0.12	0.57
1994–1997	0.42	0.72	0.58	0.75	0.46	0.76	0.28	0.77
1998–2001	0.38	1.25	−0.03	1.31	0.06	1.33	−0.20	1.35
2002–2006	0.53	1.05	0.29	1.14	0.22	1.16	0.25	1.16
Sex
Male	ref.		ref.		ref.		ref.
Female	0.09	0.42	0.20	0.49	0.25	0.50	0.30	0.51
Ethnicity
Native	ref.		ref.		ref.		ref.
Non-native	0.41	1.21	0.53	1.24	0.53	1.26	0.57	1.31
Level of education
Low	ref.		ref.		ref.		ref.
Intermediate	−0.19	0.46	−0.27	0.50	−0.36	0.51	−0.92	0.71
High	−0.79	0.64	−0.59	0.77	−0.67	0.77	−0.31	0.88
Unemployment rate	0.37*	0.18	0.33	0.19	0.32	0.19	0.32	0.19
Occupational status (first job)			−0.02	0.02	−0.01	0.02	−0.02	0.02
Type of industry (first job)
Traditional primary/classical capitalist industry			−1.41	1.21	−1.43	1.21	−1.35	1.22
Competitive industry			ref.		ref.		ref.
Large-scale engineering based industry			0.29	0.90	0.22	0.90	0.17	0.92
Small competitive industry			−0.19	0.70	−0.37	0.72	−0.43	0.73
Professional service industry			0.10	0.86	−0.09	0.88	−0.12	0.89
Bureaucratic service industry			−0.91	1.30	−0.72	1.29	−0.82	1.30
Other industry or unknown			0.32	0.87	0.25	0.88	0.19	0.90
Firm size (first job)			−0.00	0.00	−0.00	0.00	−0.00	0.00
Flexible job*Duration					−0.59*	0.25	−0.56*	0.26
Flexible job*Intermediate							1.16	1.02
Flexible job*High							−0.96	1.44
Variance	1.11	0.62	1.14	0.70	1.23	0.71	1.30	0.74

p < .01; *p < .05 (two-sided). Coefficients of ‘occupational status (first job) missing’ and ‘firm size (first job) missing’ not shown.

Source: Labour Supply Panel (1988–2008).

Model 1 in Table 2a shows that labour market entrants who start in flexible employment are more likely to have a flexible job (versus a standard one) later on in the early career than those who enter standard employment (b = 1.61). The corresponding model in Table 2b shows that flexible employment at labour market entry also enlarges the likelihood of being unemployed in the early career (b = 1.62). A flexible start in the Dutch labour market hence negatively affects the early career in terms of more employment precarity, as H1 presumed. These effects still exist after inclusion of first job characteristics, as Model 2 shows.

In Model 3, the statistical interaction term between flexible employment at labour market entry and duration since labour market entry is added. The results reveal that the penalty of a flexible entry is strongest at the start of the career (b = 2.41 with regard to flexible employment and b = 3.61 with regard to unemployment). Every additional year in the labour market decreases the penalty of starting in flexible employment by 0.26 and 0.59, respectively. In the case of flexible employment, however, the interaction term is not significant. With regard to unemployment, this finding implies that the negative effect of entering the labour market in flexible employment has vanished after about six years (3.61/0.59 = 6.12). It supports H2, which stated that the negative effects of flexible employment at labour market entry on the early career of school-leavers in the Netherlands disappear after a few years in the labour market.

In Model 4 the interactions between flexible employment at labour market entry and level of education are also estimated. None of the included interaction terms are significant. This implies that H3 cannot be corroborated.

Flexible labour market entry and occupational status development

To analyse the development in occupational status during the early career, growth curve modelling is applied. This implies that two-level linear regression models are estimated, determining the effect of flexible employment at labour market entry on occupational status attainment during the first eight years in the labour market (see Table 3).

Table 3.

Early career development in occupational status; random effects linear model (N_observations = 1831; N_respondents = 973).

	Model 1		Model 2		Model 3		Model 4		Model 5
	B	SE	B	SE	B	SE	B	SE	B	SE
Intercept	36.53**	1.65	34.87**	1.80	34.70**	1.78	34.65**	1.78	34.59**	1.81
Flexible employment (first job)
Standard job	ref.		ref.		ref.		ref.		ref.
Flexible job	−2.04**	0.78	−1.54*	0.76	−1.63*	0.76	−2.39**	0.81	−2.17	1.32
Duration	0.23*	0.11	0.24*	0.11	0.27*	0.13	0.09	0.14	0.09	0.14
Labour market entry cohort
1986–1989	ref.		ref.		ref.		ref.		ref.
1990–1993	−0.70	1.05	−0.73	1.02	−0.79	1.02	−0.77	1.02	−0.76	1.02
1994–1997	−1.84	1.11	−1.53	1.08	−1.62	1.07	−1.54	1.07	−1.54	1.08
1998–2001	0.82	1.46	0.38	1.44	0.43	1.42	0.55	1.42	0.54	1.43
2002–2006	−0.10	1.32	−0.09	1.43	−0.03	1.42	0.06	1.42	0.06	1.42
Sex
Male	ref.		ref.		ref.		ref.		ref.
Female	4.17**	0.70	2.84**	0.72	2.92**	0.72	2.95**	0.72	2.95**	0.72
Ethnicity
Native	ref.		ref.		ref.		ref.		ref.
Non-native	3.38	2.00	3.15	1.95	3.26	1.94	3.16	1.94	3.18	1.94
Level of education
Low	ref.		ref.		ref.		ref.		ref.
Intermediate	5.55**	0.75	4.40**	0.75	4.15**	0.75	4.19**	0.75	4.25**	0.87
High	17.02**	0.91	14.82**	0.96	14.60**	0.96	14.60**	0.96	14.73**	1.10
Unemployment rate	−0.06	0.19	−0.09	0.19	−0.05	0.19	−0.03	0.19	−0.03	0.19
Type of industry (first job)
Traditional primary/classical capitalist industry			−4.70**	1.75	−4.67**	1.75	−4.67**	1.75	−4.67**	1.75
Competitive industry			ref.		ref.		ref.		ref.
Large-scale engineering based industry			3.56*	1.75	3.69*	1.75	3.73*	1.75	3.70*	1.75
Small competitive industry			2.51*	1.23	2.53*	1.23	2.57*	1.23	2.56*	1.23
Professional service industry			6.49**	1.38	6.67**	1.37	6.72**	1.37	6.70**	1.38
Bureaucratic service industry			10.06**	1.69	10.23**	1.69	10.17**	1.68	10.16**	1.69
Other industry or unknown			3.64*	1.49	3.78*	1.49	3.82*	1.49	3.82*	1.49
Firm size (first job)			0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
Flexible job*Duration							0.83**	0.29	0.83**	0.29
Flexible job*Intermediate									−0.25	1.66
Flexible job*High									−0.47	1.94
Observation level
σ²	64.18**	3.105373375	63.20**	3.05	53.57**	3.05	53.61**	3.051853	53.61**	3.05
Respondent level
σ²	74.23**	5.673666429	67.72**	5.32	75.68**	6.18	75.41**	6.1644542	75.40**	6.16
σ intercept – duration					−2.88	1.51	−2.73	1.4892771	−2.72	1.49

p < .01; *p < .05 (two-sided). Coefficient of ‘firm size (first job) missing’ not shown.

Source: Labour Supply Panel (1988–2008).

Model 1 shows that starting in flexible employment coincides with a job during the early career in which, on average, less occupational status is attained (b = −2.04). Once again, this finding lends considerable support for H1. This effect largely remains when job characteristics are included (Model 2) and the slope of the duration variable is random (see Model 3).

Model 4 determines whether the initial negative effect of flexible employment disappears during the early career. At the start of the career, the effect of flexible employment on occupational status attainment is strongest (b = −2.39), but after some three years in the labour market (2.39/0.83) it has disappeared entirely and then becomes, surprisingly enough, positive.⁸ This finding gives support for H2, although the positive effect of flexible employment after three years in the labour market was not predicted.

From Model 5, it can be concluded that there are no significant differences in the effect of flexible employment at labour market entry on occupational status development during the early career between levels of education. This implies that H3, once again, cannot be supported.

Flexible labour market entry and income growth

Table 4 presents the results of the two-level linear regression models of income growth (measured in terms of standardized log net hourly wages). The results of Model 1 reveal a nearly significant negative effect of flexible employment at labour market entry on income growth during the early career (p = .13, two-sided). The same holds for Models 2 and 3. This effect becomes marginally significant (p = .08, two-sided) in Model 4, thereby giving some support for H1. At the start of the career, labour market entrants in flexible employment earn 11% less than those in standard employment. Although the initial wage disadvantage for individuals with a flexible first job seems to disappear after a few (that is, five) years, as predicted by H2, the estimated interaction term is not significant. Also the interaction terms between flexible employment at labour market entry and level of education are not significant (see Model 5). This finding gives further evidence that H3 cannot be corroborated.

Table 4.

Early career development in standardized (log) hourly wage; random effects linear model (N_observations = 1621; N_respondents = 899).

	Model 1		Model 2		Model 3		Model 4		Model 5
	B	SE	B	SE	B	SE	B	SE	B	SE
Intercept	−0.34**	0.13	−0.38**	0.14	−0.33*	0.13	−0.33*	0.13	−0.34*	0.13
Flexible employment (first job)
Standard job	ref.		ref.		ref.		ref.		ref.
Flexible job	−0.08	0.05	−0.08	0.05	−0.07	0.05	−0.11	0.06	−0.07	0.09
Duration	0.11**	0.01	0.11**	0.01	0.11**	0.01	0.11**	0.01	0.11**	0.01
Labour market entry cohort
1986–1989	ref.		ref.		ref.		ref.		ref.
1990–1993	−0.33**	0.06	−0.33**	0.06	−0.31**	0.06	−0.31**	0.06	−0.32**	0.06
1994–1997	−0.51**	0.07	−0.50**	0.07	−0.48**	0.07	−0.48**	0.07	−0.48**	0.07
1998–2001	−0.49**	0.10	−0.51**	0.10	−0.49**	0.10	−0.48**	0.10	−0.49**	0.10
2002–2006	−0.65**	0.09	−0.68**	0.10	−0.67**	0.10	−0.67**	0.10	−0.67**	0.10
Sex
Male	ref.		ref.		ref.		ref.		ref.
Female	−0.03	0.04	−0.07	0.05	−0.10	0.05	−0.10*	0.05	−0.10*	0.05
Ethnicity
Native	ref.		ref.		ref.		ref.		ref.
Non-native	0.42**	0.13	0.37**	0.13	0.32*	0.13	0.31*	0.13	0.32*	0.13
Level of education
Low	ref.		ref.		ref.		ref.		ref.
Intermediate	0.54**	0.05	0.50**	0.05	0.46**	0.05	0.46**	0.05	0.47**	0.06
High	1.26**	0.06	1.10**	0.06	1.06**	0.06	1.06**	0.06	1.08**	0.07
Unemployment rate	−0.00	0.02	0.00	0.02	−0.01	0.02	0.00	0.02	0.00	0.02
Type of industry (first job)
Traditional primary/classical capitalist industry			0.07	0.11	0.03	0.10	0.03	0.10	0.03	0.10
Competitive industry			ref.	ref.	ref.	ref.
Large-scale engineering based industry			0.09	0.11	0.07	0.11	0.08	0.11	0.07	0.11
Small competitive industry			−0.09	0.08	−0.10	0.08	−0.09	0.08	−0.09	0.08
Professional service industry			0.31**	0.09	0.27**	0.09	0.28**	0.09	0.28**	0.09
Bureaucratic service industry			0.22*	0.11	0.21*	0.10	0.21*	0.10	0.21*	0.10
Other industry or unknown			0.21*	0.09	0.18*	0.09	0.19*	0.09	0.19*	0.09
Firm size (first job)			0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
Flexible job*Duration							0.02	0.02	0.02	0.02
Flexible job*Intermediate									−0.03	0.11
Flexible job*High									−0.09	0.13
Observation level
σ²	0.44**	0.02	0.44**	0.02	0.36**	0.02	0.37**	0.02	0.37**	0.02
Respondent level
σ²	0.16**	0.03	0.14**	0.02	0.29**	0.04	0.29**	0.04	0.29**	0.04
σ intercept – duration					−0.04**	0.01	−0.04**	0.01	−0.04**	0.01

p < .01; * p < .05 (two-sided). Coefficient of ‘firm size (first job) missing’ not shown.

Conclusion and discussion

The empirical results of this article unambiguously lead to the conclusion that school-leavers who enter the Dutch labour market in a flexible job are more likely to stay in flexible employment, become unemployed, attain lower occupational status and earn less income in the early career than those who start in a standard job. The detrimental effects of flexible employment at labour market entry, however, are only temporary and diminish after some years in the labour market. Although not all relevant estimates are significant, the results unequivocally go into the same, expected direction. So, we can be fairly confident of the reality of scar effects of flexible employment, but the findings undermine the entrapment hypothesis, which claims long-lasting, negative career consequences of a flexible start. Instead, the stepping-stone hypothesis is more convincing, given that the negative effects of flexible employment at labour market entry on the early career of school-leavers are persistent, but not permanent.

One further aim of this article was to study the predicted differential effect of education with regard to the early career effects of flexible employment at labour market entry among school-leavers in the Netherlands. In the empirical analysis, however, no single indication was found for this differential effect. In fact, the negative consequences of flexible employment at labour market entry for the early career do not differ between lower and higher educated school-leavers. Apparently, in the Netherlands, long-term risks and uncertainties in the career are not directly transferred to the more vulnerable, lower educated part of the labour force in particular. Although the message that stems from these results is rather neutral – that is, concerns that education might operate as a contributor to cumulative labour market disadvantages cannot be confirmed, nor that education (alone) is able to protect from the negative consequences of early flexible employment – one has to be careful in concluding that level of education is not relevant when entering the labour market in a flexible job. In fact, level of education strongly determines the likelihood of starting in a flexible job, but the effect of flexible employment at labour market entry on the early career mediates the role of level of education and, moreover, is not moderated by it.

The conclusion that the results in this article point towards the integration rather than the entrapment scenario corresponds to findings from comparable studies on other European countries (see, for instance, Gash, 2008). Our results, however, contrast previous findings for the Netherlands (Steijn et al., 2006; Wolbers, 2010) that suggested that flexible employment at labour market entry seems to have long-lasting, negative consequences for the further career. Apart from differences in the design – in this article we used a prospective panel study, whereas the earlier studies analysed cross-sectional or retrospective data, in which recall bias may have occurred – these contradictory findings for the Netherlands may be related to a different treatment of unobserved heterogeneity. Although there is really no simple way to tackle this issue, we tried to remove any possible bias that results from the endogeneity of the labour market entry position in this article, by including specific covariates in order to avoid at least most of the selection problem. In addition to controlling for important sociodemographic characteristics of individuals that largely determine their opportunities in the labour market (level of education, gender, ethnicity and so on), we used the aggregate unemployment rate as an exogenous labour market indicator to get around the endogeneity problem. Nevertheless, it remains possible that some confounding factors have still been left out, thereby overestimating the true scar effects of flexible employment. Given that the findings suggest that the effects of flexible employment at labour market entry fade during the early career, we do not, however, consider the possible bias to be a crucial objection to our analysis (see also Skans [2011], who finds a similar time pattern in an analysis of unemployment scarring and makes an argument in this direction).

A final remark concerns the measurement of flexible employment. Although various types of flexible employment (temporary contracts, temporary agency work, on-call employment and so on) have shown not to be equally detrimental for school-leavers when entering the Dutch labour market (de Vries and Wolbers, 2005), it was not possible to distinguish between types of flexible employment in the empirical analysis of this article. First of all, the classification of the various types of flexible employment differed between the moments of interview, which forced us to pool all types together into one category (on the basis of the lowest common denominator). Second, given the relatively small sample sizes of school-leavers, it is very likely that the statistical power would have been too low to detect any statistically significant differences between the various types of flexible employment. Nevertheless, future research should focus on solving these issues in order to find out whether some types of flexible employment at labour market entry are less detrimental for the early career of school-leavers than others.

Footnotes

Declaration of conflicting interests

The author declares that there is no conflict of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

Author biographies

Marloes de Lange is a PhD student in the Department of Sociology, Radboud University Nijmegen, the Netherlands. Currently, she is writing a dissertation on labour market flexibilization among young people in the Netherlands.

Maurice Gesthuizen is an assistant professor in the Department of Sociology, Radboud University Nijmegen, the Netherlands. He is interested in educational inequality, economic vulnerability, social capital and their interrelationships. He studies these topics in longitudinal and comparative perspective.

Maarten HJ Wolbers is an associate professor in the Department of Sociology, Radboud University Nijmegen, the Netherlands. He previously worked at the Research Centre for Education and the Labour Market, Maastricht University and the Department of Social Research Methodology, VU University Amsterdam. His research interests include the sociology of education, work and labour markets.

References

Arulampalam

Gregg

Gregory

(2001) Unemployment scarring. The Economic Journal 111(475): F577–F584.

Baranowska

Gebel

Kotowska

(2011) The role of fixed-term contracts at labour market entry in Poland: Stepping stones, screening devices, traps or search subsidies? Work, Employment and Society 25(4): 777–793.

Baron

Bielby

(1984) The organization of work in a segmented economy. American Sociological Review 49: 454–473.

Becker

(1964). Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education. New York: National Bureau of Economic Research.

Blossfeld

H-P

Buchholz

Bukodi

Kurz

(eds) (2008) Young Workers, Globalization and the Labor Market: Comparing Early Working Life in Eleven Countries. Cheltenham, UK and Northampton, MA: Edward Elgar.

Bratberg

Nilsen

(2000) Transitions from school to work and the early labour market experience. Oxford Bulletin of Economics and Statistics 62 (Suppl.): 909–929.

Breen

(1997) Risk, recommodification and the future of the service class. Sociology 31(3): 473–489.

CBS (2011) StatLine databank. Voorburg/Heerlen: Centraal Bureau voor de Statistiek. Available at: statline.cbs.nl (accessed 3 August 2011).

De Graaf-Zijl

Van den Berg

Heyma

AOJ

(2011) Stepping stones for the unemployed: The effect of temporary jobs on the duration until (regular) work. Journal of Population Economics 24(1): 107–139.

10.

De Jong

De Cuyper

De Witte

. (2009) Motives for accepting temporary employment: A typology. International Journal of Manpower 30(3): 237–252.

11.

De Lange

Gesthuizen

Wolbers

MHJ

(2012) Trends in labour market flexibilization among Dutch school-leavers: The impact of economic globalization on educational differences. International Sociology 27(4): 529–550.

12.

De Vries

Wolbers

MHJ

(2005) Non-standard employment relations and wages among school-leavers in the Netherlands. Work, Employment and Society 19(3): 503–525.

13.

Doeringer

Piore

(1971) Internal Labour Markets and Manpower Analysis. Lexington, MA: Heath.

14.

Gangl

(2006) Scar effects of unemployment: An assessment of institutional complementarities. American Sociological Review 71(6): 986–1013.

15.

Ganzeboom

HBG

De Graaf

Treiman

(1992) A standard international socio-economic index of occupational status. Social Science Research 21(1): 1–56.

16.

Gash

(2008) Bridge or trap? Temporary workers’ transitions to unemployment and to the standard employment contract. European Sociological Review 24, 651–68.

17.

Gebel

(2010) Early career consequences of temporary employment in Germany and the UK. Work, Employment and Society 24(5): 641–660.

18.

Golsch

(2003) Employment flexibility in Spain and its impact on transitions to adulthood. Work, Employment and Society 17(4): 691–718.

19.

Goslinga

Sverke

(2003) Atypical work and trade union membership: Union attitudes and union turnover among traditional vs. atypically employed union members. Economic and Industrial Democracy 24(2): 290–312.

20.

Iannelli

Soro-Bonmati

(2003) Transition pathways in Italy and Spain: Different patterns, similar vulnerability? In: Müller

Gangl

(eds) Transitions from Education to Work in Europe: The Integration of Youth into EU Labour Markets. Oxford: Oxford University Press, pp. 212–250.

21.

Kalleberg

(2009) Precarious work, insecure workers: Employment relations in transition. American Sociological Review 74(1): 1–22.

22.

Kalleberg

Reskin

Hudson

(2000) Bad jobs in America: Standard and nonstandard employment relations and job quality in the United States. American Sociological Review 65(2): 256–278.

23.

Katz

Autor

(1999) Changes in the wage structure and earnings inequality. In: Ashenfelter

Card

(eds) Handbook of Labour Economics, Vol. 3. Amsterdam: North-Holland, pp. 1463–1555.

24.

Kogan

Müller

(2003) School-to-Work Transitions in Europe. Mannheim: Mannheimer Zentrum für Europäische Sozialforschung.

25.

Luijkx

Wolbers

MHJ

(2009) The effects of non-employment in early work-life on subsequent employment chances of individuals in the Netherlands. European Sociological Review 25(6): 647–660.

26.

McGinnity

Mertens

Gundert

(2005) A bad start? Fixed-term contracts and the transition from education to work in West-Germany. European Sociological Review 21(4): 359–374.

27.

Müller

Gangl

(2003) Transitions from Education to Work in Europe: The Integration of Youth into EU Labour Markets. Oxford: Oxford University Press.

28.

OECD (2004) Employment Outlook. Paris: Organization for Economic Co-operation and Development.

29.

Remery

Van Doorne-Huiskes

Schippers

(2002) Labour market flexibility in the Netherlands: Looking for winners and losers. Work, Employment and Society 16(3): 477–495.

30.

Scherer

(2004) Stepping-stones or traps? The consequences of labour market entry positions on future careers in West-Germany, Great Britain and Italy. Work, Employment and Society 18(2): 369–394.

31.

Scherer

(2005) Patterns of labour market entry: Long wait or career instability? An empirical comparison of Italy, Great Britain and West-Germany. European Sociological Review 21(5): 427–440.

32.

Skans

(2011) Scarring effects of the first labor market experience. IZA Discussion Paper No. 5565. Institute for the Study of Labour, Bonn.

33.

Spence

(1973) Job market signaling. Quarterly Journal of Economics 87(3): 355–374.

34.

Steijn

Need

Gesthuizen

(2006) Well begun, half done? Long-term effects of labour market entry in the Netherlands, 1950–2000. Work, Employment and Society 20(3): 453–472.

35.

Stinchcombe

(1979) Social mobility in industrial labour markets. Acta Sociologica 22(3): 217–245.

36.

Visser

(2002) The first part-time economy in the world: A model to be followed? Journal of European Social Policy 12(1): 23–42.

37.

Visser

Hemerijck

(1997) A Dutch Miracle: Job Growth, Welfare Reform and Corporatism in the Netherlands. Amsterdam: Amsterdam University Press.

38.

Wang

Weiss

(1998) Probation, layoffs, and wage-tenure profiles: A sorting explanation. Labour Economics 5(3): 359–383.

39.

Wolbers

MHJ

(2008) Increasing labor market instability among young people? Labor market entry and early career development among school-leavers in the Netherlands since the mid-1980s. In: Blossfeld

H-P

Bucholz

Bukodi

Kurz

(eds) Young Workers, Globalization and the Labor Market: Comparing Early Working Life in Eleven Countries. Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 77–101.

40.

Wolbers

MHJ

(2010) Blijvende gevolgen van een gebrekkige start? Tijdschrift voor Arbeidsvraagstukken 26: 223–239.